Plankton response to global warming is characterized by non-uniform shifts in assemblage composition since the last ice age

Abstract

Biodiversity is expected to change in response to future global warming. However, it is difficult to predict how species will track the ongoing climate change. Here we use the fossil record of planktonic foraminifera to assess how biodiversity responded to climate change with a magnitude comparable to future anthropogenic warming. We compiled time series of planktonic foraminifera assemblages, covering the time from the last ice age across the deglaciation to the current warm period. Planktonic foraminifera assemblages shifted immediately when temperature began to rise at the end of the last ice age and continued to change until approximately 5,000 years ago, even though global temperature remained relatively stable during the last 11,000 years. The biotic response was largest in the mid latitudes and dominated by range expansion, which resulted in the emergence of new assemblages without analogues in the glacial ocean. Our results indicate that the plankton response to global warming was spatially heterogeneous and did not track temperature change uniformly over the past 24,000 years. Climate change led to the establishment of new assemblages and possibly new ecological interactions, which suggests that current anthropogenic warming may lead to new, different plankton community composition.

Coherent response of zoo‐ and phytoplankton assemblages to global warming since the Last Glacial Maximum

“…to climate change over the past 24 kyr with previously established trends in planktonic foraminifera (Strack et al, 2022). The joint analyses of the assemblage composition time series of these three plankton groups reveal that the transformation of the marine envi- (Figures 5a and 6).…”

“…The analyses are based on a previous compilation of 25 planktonic foraminifera assemblage time series (Strack et al, 2022) that, after an exhaustive search, has been expanded with 6 organic-walled dinoflagellate cyst (dinocyst) and 6 coccolithophore assemblage time series to compare patterns across groups (see Supporting Information Table S1 for full list of time series). Throughout this study, the terms 'assemblage' and 'community' are used as defined by Fauth et al (1996), where 'community' refers to all species that occur in the same place at the same time, and 'assemblage' refers to all taxa of phylogenetically related groups within a community.…”

“…To ensure the chronological comparability of all analysed time series, we either used the revised age models included in the PALMOD 130k marine palaeoclimate data synthesis V1.1 (Jonkers et al, 2020) or we revised the published age models following the same approach as in Jonkers et al (2020). Their age models are based on benthic foraminifera oxygen isotope data that have been manually calibrated to regional benthic foraminifera oxygen stacks (Lisiecki & Stern, 2016) as well as radiocarbon ages (see also Strack et al, 2022).…”

“…we followed the analysis of Strack et al (2022) where S is the number of species at a specific site and p i is the relative abundance of the ith species. for a more detailed explanation).…”

“…It is important to note that when expressed in this way, the slopes only approximate the overall amount of change across time, and the underlying change may not have been monotonic. In this approximation, negative trends in certain parameters indicate that younger samples in the time series are less different from the baseline (the oldest sample in the time series) than the older samples.Positive trends indicate that younger samples in the time series are more different from the baseline than the older samples (see Extended Data Figures3 and 4and Methods ofStrack et al (2022)…”